flowchart
subgraph Research
direction LR
M(Model) --> I(Inquiry) --> D(Data) --> A(Analysis) --> R(Remodel) --> M(Model)
end
P{Puzzle} -.-> Research
style Research fill:#282828,stroke:#fbf1c7,stroke-width:1px,color:#689d6a
style P fill:#282828,stroke:#fbf1c7,stroke-width:2px,color:#689d6a,font-size:20px
classDef largeNode fill:#282828,stroke:#fbf1c7,stroke-width:2px,color:#689d6a,font-size:20px
class M,I,D,A,R largeNode
What is political science,
and how do we do scientific research?
PSCI 2270 - Lecture 1
August 29, 2024
Science is curiosity
Even more in reality
Looking for the Question
- Personal experiences and histories!
News/Blogs:
- Books, movies/TV series (e.g. Borgen, Veep, the Wire, etc.)
“I have stood on the shoulders of giants” — Isaac Newton
- ⛏️: Vanderbilt Library, Google scholar
- 💻: Elicit, Semantic Scholar, SCite,…
- 🤖: ChatGPT or Claude are NOT good at creating new questions/reviewing literature
- ResearchRabbit for literature networks
- Even social media: YouTube, Twitter ( \(\mathbb{X}\) ), TikTok, etc.
Good models cont’d
Simplified picture of the world
- Theoretically
- Observationally
Use Occam’s razor
- Prefer parsimony
- Explain a lot with a little
“All models are wrong, but some are useful.” — George Box
Useful Wrong Model
Kenworthy and Pontusson (2005)
In a 2005 study, Lane Kenworthy and Jonas Pontusson analyzed trends in the distribution of gross market income-the distribution of income before taxes and government transfers-for affluent Organisation for Economic Co-operation and Development (OECD) countries using data from the Luxembourg Income Study. Kenworthy and Pontusson were interested in whether inequality in market income had increased and to what extent government policies had responded to changes in market income inequality. In particular, they were interested in testing the median-voter model developed by Allan H. Meltzer and Scott F Richards…
What if we consider turnout?
Mendelberg, Karpowitz, and Oliphant (2014)
Is the number of women in a legislative body the critical factor [for substantive representation], or might the rules governing deliberation in the legislature also be important?.. Mendelberg, Karpowitz, and Oliphant’s research investigates whether [this relationship is] affected by a group’s decision rule. They hypothesize that under a unanimous rule, women will receive more respect in deliberations… but only when women are in the minority, not when they predominate (based on previous research). To test their hypothesis, they set up 94 five-member discussion groups composed of between O and 5 women, and randomly assigned each group to unanimous or majority rule. The.researchers … counted the number of times each person spoke and coded the number and tone (positive, neutral, or negative) of interruptions…
Butler and Broockman (2011)
We use a field experiment to investigate whether race affects how responsive state legislators are to requests for help with registering to vote. In an email sent to each legislator, we randomized whether a putatively black or white alias was used and whether the email signaled the sender’s partisan preference. Overall, we find that putatively black requests receive fewer replies. We explore two potential explanations for this discrimination: strategic partisan behavior and the legislators’ own race. We find that the putatively black alias continues to be differentially treated even when the emails signal partisanship, indicating that strategic considerations cannot completely explain the observed differential treatment. Further analysis reveals that white legislators of both parties exhibit similar levels of discrimination against the black alias. Minority legislators do the opposite, responding more frequently to the black alias. Implications for the study of race and politics in the United States are discussed.
Aytaç, Schiumerini, and Stokes (2017)
Elected governments sometimes deal with protests by authorizing the police to use less-lethal tools of repression: water cannons, tear gas, rubber bullets, and the like. When these tactics fail to end protests and instead spark larger, backlash movements, some governments reduce the level of violence but others increase it, causing widespread injuries and loss of life. We study three recent cases of governments in new democracies facing backlash movements. Their decision to scale up or scale back police repression reflected the governments’ levels of electoral security… Our original survey research and interviews with civilian authorities, police officials, and protest organizers in Turkey, Brazil, and Ukraine allow us to evaluate this explanation as well as a number of rival accounts. Our findings imply that elected governments that rest on very stable bases of support may be tempted to deploy tactics more commonly associated with authoritarian politics.
Broockman and Kalla (2024)
Many Americans consume aligned partisan media, which scholars worry contributes to polarization. Many propose encouraging these Americans to consume cross-cutting media to moderate their attitudes. However, motivated reasoning theory posits that exposure to crosscutting media could backfire, exacerbating polarization. Building on theories that sustained exposure to novel information can overcome motivated reasoning and that partisan sources on opposite sides cover distinct information, we argue that sustained consumption of crosscutting media leads voters to learn uncongenial information and moderate their attitudes in covered domains. To test this argument, we used data on actual TV viewership to recruit a sample of regular Fox News viewers and incentivized a randomized treatment group to watch CNN instead for a month. Contrary to predictions from motivated reasoning, watching CNN caused substantial learning and moderated participants’ attitudes in covered domains.